V0.1.2

.github/workflows/cd.yml

@@ -0,0 +1,33 @@
+name: CD
+
+on:
+  pull_request:
+    types: [closed]
+    branches:
+      - main
+
+jobs:
+  build-n-publish:
+    runs-on: ubuntu-latest
+    steps:
+      - uses: actions/checkout@v2
+
+      - name: Set up Python 3.10
+        uses: actions/setup-python@v2
+        with:
+          python-version: 3.10
+
+      - name: Install Dependencies
+        run: |
+          python -m pip install --upgrade pip
+          pip install tox
+
+      - name: Test With Tox
+        run: tox -e pytest
+
+      - name: Build and Publish
+        env:
+          TWINE_USERNAME: __token__
+          TWINE_PASSWORD: ${{ secrets.PYPI_API_TOKEN }}
+        run: |
+          tox -r -e build,publish

.github/workflows/tests.yml → .github/workflows/ci.yml

@@ -44,11 +44,11 @@ jobs:
         run: |
           make lint
           make lint-notebooks
-        continue-on-error: true
+        continue-on-error: false
 
-      - name: Run Tests with Tox
+      - name: Run Tests and Coverage
         run: |
-          make tox
+          make coverage
 
       # TO DO: Add codecov
       # - name: Upload coverage to Codecov

Makefile

@@ -8,13 +8,15 @@ PYLINT := pylint
 TOX := $(PYTHON) -m tox
 SETUP := $(PYTHON) setup.py
 
+DIR := denseclus
+
 .PHONY: lint lint-notebooks test coverage tox install install-dev pypi clean help
 
 lint:
 	@echo "Running linting..."
-	@$(BLACK) denseclus tests setup.py
-	@$(RUFF) denseclus tests setup.py --fix --preview
-	@$(PYLINT) denseclus --disable=R0902,W0222,W0221,C0103,W0632
+	@$(BLACK) $(DIR) tests setup.py
+	@$(RUFF) $(DIR) tests setup.py --fix --preview
+	@$(PYLINT) $(DIR) --disable=R0902,W0222,W0221,C0103,W0632
 
 lint-notebooks:
 	@echo "Linting notebooks..."
@@ -29,7 +31,7 @@ test:
 coverage:
 	@echo "Running coverage..."
 	@$(COVERAGE) erase
-	@$(COVERAGE) run --include=denseclus/* -m pytest -ra
+	@$(COVERAGE) run --source=$(DIR) -m pytest -ra
 	@$(COVERAGE) report -m
 
 tox:

denseclus/DenseClus.py

@@ -180,14 +180,14 @@ def __init__(
             self.umap_params = default_umap_params
 
         if hdbscan_params:
-            self.hdbscan_params = hdbscan_params
+            self.hdbscan_params = hdbscan_params  # pragma: no cover
         else:
             self.hdbscan_params = default_hdbscan_params
 
-        if verbose:
+        if verbose:  # pragma: no cover
             logger.setLevel(logging.DEBUG)
             self.verbose = True
-        else:
+        else:  # pragma: no cover
             logger.setLevel(logging.ERROR)
             self.verbose = False
 
@@ -201,13 +201,21 @@ def noop(*args, **kargs):
 
         if isinstance(random_state, int):
             np.random.seed(seed=random_state)
-        else:
+        else:  # pragma: no cover
             logger.info("No random seed passed, running UMAP in Numba, parallel")
 
         self.kwargs = kwargs
 
-    def __repr__(self):
-        return str(self.__dict__)
+    def __repr__(self):  # pragma: no cover
+        return f"""DenseClus(random_state={self.random_state}
+                            ,umap_combine_method={self.umap_combine_method}
+                            ,umap_params={self.umap_params}
+                            ,hdbscan_params={self.hdbscan_params})"""
+
+    def __str__(self):  # pragma: no cover
+        return f"""DenseClus object (random_state={self.random_state}
+                                    ,umap_combine_method={self.umap_combine_method}
+                                    ,verbose={self.verbose})"""
 
     def fit(self, df: pd.DataFrame) -> None:
         """Fits the UMAP and HDBSCAN models to the provided data.

denseclus/__init__.py

@@ -8,7 +8,7 @@
 from .DenseClus import DenseClus
 from .utils import extract_categorical, extract_numerical
 
-__version__ = "0.1.1"
+__version__ = "0.1.2"
 
-if __name__ == "__main__":
+if __name__ == "__main__":  # pragma: no cover
     print(type(DenseClus), type(extract_categorical), type(extract_numerical))

denseclus/utils.py

@@ -37,8 +37,6 @@ def extract_categorical(
         raise TypeError("Input should be a pandas DataFrame")
     if df.empty:
         raise ValueError("Input DataFrame should not be empty.")
-    if df.shape[1] < 1:
-        raise ValueError("Input DataFrame should have at least one column.")
 
     categorical = df.select_dtypes(exclude=["float", "int"])
 
@@ -128,8 +126,6 @@ def extract_numerical(df: pd.DataFrame, impute_strategy: str = "median", **kwarg
         raise TypeError("Input should be a pandas DataFrame")
     if df.empty:
         raise ValueError("Input DataFrame should not be empty.")
-    if df.shape[1] < 1:
-        raise ValueError("Input DataFrame should have at least one column.")
 
     numerical = df.select_dtypes(include=["float", "int"])
     if numerical.shape[1] == 0:

pyproject.toml

@@ -84,10 +84,10 @@ exclude = [
 ]
 
 [tool.pytest.ini_options]
-addopts = "--cov=src --cov-report=term-missing"
+addopts = "--cov=denseclus --cov-report=term-missing"
 markers = [
-    "unit",
-    "integration"
+    "fast",
+    "slow"
 ]
 
 [tool.pytest]
@@ -96,6 +96,9 @@ doctest_optionflags = [
     "ELLIPSIS",
     "NUMBER",
 ]
+filterwarnings = [
+   "ignore::UserWarning"
+]
 
 [tool.pylint.'MESSAGES CONTROL']
 disable = [

setup.py

@@ -9,7 +9,7 @@
 
 setuptools.setup(
     name="amazon_denseclus",
-    version="0.1.1",
+    version="0.1.2",
     author="Charles Frenzel & Baichuan Sun",
     description="Dense Clustering for Mixed Data Types",
     long_description=long_description,

tests/utils_test.py

@@ -172,3 +172,13 @@ def test_extract_numerical_columns(df, numerical_df):
     # Test with DataFrame with no numeric columns
     with pytest.raises(ValueError):
         extract_numerical(pd.DataFrame({"col1": ["a", "b", "c"]}))
+
+
+def test_impute_categorical_strategy_error(categorical_df):
+    with pytest.raises(ValueError):
+        impute_categorical(categorical_df, strategy="median", fill_value="Error")
+
+
+def test_impute_numerical_strategy_error(numerical_df):
+    with pytest.raises(ValueError):
+        impute_numerical(numerical_df, strategy="constant")

tox.ini

@@ -4,24 +4,20 @@
 [main]
 line_len = 100
 src_dir =
-    notebooks
     denseclus
 
 [tox]
 isolated_build = True
 minversion = 3.15
 envlist = py310
-
+requires = tox>=3.15
 
 [testenv]
 deps =
     pytest
     pytest-cov
-    ruff
-    black
     coverage
     mypy
-    pylint
 
 description = invoke pytest to run automated tests
 ;install_command = pip install {opts} {packages}
@@ -32,13 +28,9 @@ passenv =
 extras =
     testing
 commands =
-    pytest --cov=src --cov-report=term-missing {posargs}
-    black denseclus
-    ruff denseclus --fix --preview
-    pylint --fail-under=7 denseclus --disable=R0902,W0222,W0221,C0103
-    mypy denseclus
-    coverage erase
-    coverage run --include=denseclus/* -m pytest -ra
+    pytest {posargs}
+    mypy {[main]src_dir}
+    coverage run --include={[main]src_dir}/* -m pytest -ra
     coverage report -m
 
 
@@ -90,43 +82,17 @@ commands =
     python -m twine upload {posargs:--repository pypi} dist/*
 
 
-[mypy]
-ignore_missing_imports = True
-files = **/*.py
-
 [pytest]
 addopts =
     -v
 filterwarnings =
 # https://github.com/boto/boto3/issues/1968
     ignore:Using or importing the ABCs from 'collections' instead of from 'collections.abc' is deprecated
 
-[testenv:ruff]
-description = check ruff formatting for CI
-deps =
-    ruff
-commands =
-    ruff --max-complexity=20 --exclude=build/, .git, __pycache__, .*_cache, examples/, .tox, data/, venv/, .venv/ --ignore=E203,W503,E722,E231 {[main]src_dir}
 
-[testenv:isort-check]
-description = check ruff formatting for CI
+[testenv:lint]
 deps = ruff
-commands =
-    ruff --line-length=100 --check --atomic --profile=black {[main]src_dir}
-
-
-[testenv:black-check]
-description = check black formatting for CI
-deps = black==23.11.0
-commands =
-    black -l {[main]line_len} --check {[main]src_dir}
-
-[testenv:mypy]
-deps =
-    mypy
-commands =
-    mypy --config-file tox.ini {[main]src_dir}
-
+commands = ruff --max-complexity=20 --exclude=build/, .git, __pycache__, .*_cache, examples/, .tox, data/, venv/, .venv/ --ignore=E203,W503,E722,E231 {[main]src_dir}
 
 [testenv:pytest]
 deps =
@@ -138,3 +104,16 @@ commands =
 [gh-actions]
 python =
     3.10: py310
+
+
+[testenv:mypy]
+deps =
+    mypy
+commands =
+    mypy --config-file tox.ini {[main]src_dir}
+ignore_missing_imports = True
+files = {[main]src_dir}/*.py
+
+[mypy]
+ignore_missing_imports = True
+files = {[main]src_dir}/*.py